Kimia Lab, University of Waterloo, Waterloo, ON, N2L 3G1, Canada.
Vector Institute, MaRS Centre, Toronto, ON, M5G 1M1, Canada.
Sci Rep. 2021 May 10;11(1):9817. doi: 10.1038/s41598-021-89194-4.
Fast diagnosis and treatment of pneumothorax, a collapsed or dropped lung, is crucial to avoid fatalities. Pneumothorax is typically detected on a chest X-ray image through visual inspection by experienced radiologists. However, the detection rate is quite low due to the complexity of visual inspection for small lung collapses. Therefore, there is an urgent need for automated detection systems to assist radiologists. Although deep learning classifiers generally deliver high accuracy levels in many applications, they may not be useful in clinical practice due to the lack of high-quality and representative labeled image sets. Alternatively, searching in the archive of past cases to find matching images may serve as a "virtual second opinion" through accessing the metadata of matched evidently diagnosed cases. To use image search as a triaging or diagnosis assistant, we must first tag all chest X-ray images with expressive identifiers, i.e., deep features. Then, given a query chest X-ray image, the majority vote among the top k retrieved images can provide a more explainable output. In this study, we searched in a repository with more than 550,000 chest X-ray images. We developed the Autoencoding Thorax Net (short AutoThorax -Net) for image search in chest radiographs. Experimental results show that image search based on AutoThorax -Net features can achieve high identification performance providing a path towards real-world deployment. We achieved 92% AUC accuracy for a semi-automated search in 194,608 images (pneumothorax and normal) and 82% AUC accuracy for fully automated search in 551,383 images (normal, pneumothorax and many other chest diseases).
快速诊断和治疗气胸（肺部塌陷或萎陷）对于避免死亡至关重要。气胸通常通过经验丰富的放射科医生进行视觉检查在胸部 X 光图像上检测到。然而，由于小范围肺塌陷的视觉检查非常复杂，因此检测率相当低。因此，迫切需要自动化检测系统来协助放射科医生。尽管深度学习分类器在许多应用中通常能提供高精度水平，但由于缺乏高质量和有代表性的标记图像集，它们在临床实践中可能并不有用。或者，通过访问匹配诊断病例的元数据，可以在过去病例档案中搜索以找到匹配的图像，作为“虚拟第二意见”。为了将图像搜索用作分诊或诊断辅助，我们必须首先使用有表现力的标识符（即，深度特征）标记所有胸部 X 光图像。然后，对于查询胸部 X 光图像，在检索到的前 k 个图像中进行多数投票，可以提供更具可解释性的输出。在这项研究中，我们在一个拥有超过 550,000 张胸部 X 光图像的存储库中进行了搜索。我们开发了用于胸部 X 射线图像的自动编码胸廓网络（简称 AutoThorax-Net）进行图像搜索。实验结果表明，基于 AutoThorax-Net 特征的图像搜索可以实现高识别性能，为实际应用提供了一条途径。我们在 194,608 张图像（气胸和正常）中实现了半自动搜索的 92% AUC 准确率，在 551,383 张图像（正常、气胸和许多其他胸部疾病）中实现了全自动搜索的 82% AUC 准确率。
